Characteristics and Needs of Psychiatric Patients With Prolonged Hospital Stay

Abstract

Objective:

To describe the characteristics and needs prior to, on admission, during the first month in hospital, at the thirtieth day of hospitalization and posthospital discharge of psychiatric patients occupying acute beds.

Methods:

This prospective observational study was conducted in 2 tertiary care hospitals. Adult patients hospitalized on a psychiatric unit for 30 days were identified. Data was collected from their medical charts and interviews with their health care team. The categorization of acute and nonacute status at day 30 was based on the health care professional’s evaluation. Descriptive and univariate analyses were performed.

Results:

A total of 262 patients were identified (mean age 45 years), 66% lived at home and 11% were homeless. More than one-half were cognitively impaired and a few had special medical needs. Ninety-seven per cent had been admitted from the emergency department. At day 30, 81% of patients required acute care, while 19% (95% CI 15% to 24%) occupied an acute care bed, despite the resolution of their acute condition. The main reason preventing discharge of nonacute patients was the difficulty or inability to find appropriate resources that met patients’ needs. As for patients who required acute care, the most common psychiatric issues were delusions or hallucinations (34%), inability to take medications independently (23.6%), and inadequate control of aggression or impulsivity (16.5%).

Conclusions:

Prevention of the discharge of nonacute patients is largely due to the difficulty in finding appropriate resources that meet patients’ needs. Improved access to community and subacute care resources could potentially facilitate the hospital discharge of psychiatric nonacute patients.

It is recognized that outpatient resources for patients who suffer from psychiatric illnesses are insufficient.^1–4 As a result, a significant number of psychiatric patients use EDs as their main entrance into the health care system^5–8 and as a source of primary care.^2,3,9–11 Due to their complex needs,^2,9,12 a significant percentage of psychiatric patients presenting to the ED require inpatient care. This is reflected by their high admission rates (22% to 35%^2,8,13–15; and up to 54% for patients with substance abuse).¹⁵ This high admission rate, combined with a lack of acute care beds on inpatient psychiatric units^1–3,10,11 results in boarding of admitted patients in ED corridors.

It is accepted that acute care resources, including hospital beds, should be used by patients who need them the most.¹⁶ With limited acute care psychiatry beds to adequately meet the needs, it becomes imperative that patients not remain in hospital after their acute care episode resolves. However, due to the specific needs of this clientele, patients are not leaving the acute care psychiatry units in a timely manner.

Lack of housing and community support for patients requiring long-term care is commonly cited among the reasons that prevent hospital discharges and extend hospital stays on acute psychiatry units.¹⁷ As a result, an important proportion of acute psychiatry beds are occupied inappropriately because of the absence of suitable alternative settings.¹⁸ For example, Paton et al¹⁹ reported that up to 40% of elderly patients were inappropriately hospitalized on an acute psychiatry unit, primarily because no other alternatives were available. This is not a new problem; more than 15 years ago, the major reasons preventing hospital discharge from acute psychiatry wards included a lack of suitable alternatives, such as long-term rehabilitation beds or domiciliary based community support.²⁰

Clinical Implications

• It is not uncommon for admitted psychiatric patients to board in the ED for lengthy periods of time due to difficulties in accessing acute hospital psychiatric beds.

• Psychiatry beds are limited and consequently, the demand for acute hospital care is not adequately addressed. Therefore, it is imperative to identify the reasons that prevent the discharge of nonacute patients.

• The obstacle to discharging nonacute patients is largely the difficulty in finding appropriate resources that meet patient needs. Improved access to community and subacute care resources could potentially facilitate the hospital discharge of psychiatric nonacute patients.

Limitations

• The categorization of acute and nonacute status was based on the subjective opinions of the health care professionals managing the patients as specified in the patients’ medical charts.

• Our study does not address the needs of nonacute patients discharged prior to 30 days of hospitalization.

• The small sample of patients categorized as nonacute in our study prevents us from identifying factors related to inappropriate use of acute care beds.

Previous studies^17,21–24 have also shown that older age, cognitive impairment,^23,24 functional dependence,²³ as well as the presence of multiple comorbidities and physical illnesses²⁵ were associated with longer hospital LOS for psychiatric patients. Additionally, diagnoses of schizophrenia,^26,27 psychotic or mood disorders,^21,22 and the presence of aggression or agitation²⁶ are associated with longer hospital stays.

Our study’s objective was to describe psychiatric patients occupying acute beds in a psychiatric unit and based on their needs, to determine the level of care required (acute, compared with nonacute) at day 30 of hospitalization. Specific objectives included describing patients’ characteristics and needs during the following times: prior to hospitalization, on admission, during the first month in hospital, at the thirtieth day of hospitalization, and posthospital discharge.

Methods

Study Design and Settings

Our prospective observational study was conducted in 2 university tertiary care hospitals in Montreal from March 2009 to April 2010. The JGH has 637 hospital beds, of which 48 are psychiatry beds, while NDH has 403 inpatient beds, of which 56 are psychiatry beds. In 2009–2010, admissions to a psychiatry unit in both hospitals represented 2.9% and 6.7% of their total hospitalizations, respectively. Our study was approved by the Research Ethics Committees of both hospitals. Since the study did not involve patient contact, no written informed consent was required.

Participants

Adult patients (18 years or older) admitted to the psychiatry unit were identified on the thirtieth day (plus or minus 3 days) of their hospitalization. Patients with psychiatric problems not admitted to the psychiatry unit were excluded. Day 30 was selected first because it closely represents the average hospital LOS on psychiatric units at both hospitals (2009–2010 = 29.6 days). Second, according to the latest Quebec provincial data, patients admitted to an acute psychiatry unit in Montreal stayed in hospital on average 31.5 days.²⁸

Data Collection

Data were collected from chart reviews and additional information was obtained when needed from the appropriate health care professionals. Five categories of data were collected: pre-admission (for example, sociodemographics, living arrangements, psychosocial condition, and special medical needs), on admission (for example, type of admission and admission diagnosis), from admission to day 30 (for example, consultations, evaluations, and treatments), on day 30 (for example, current diagnosis, active psychiatric problems, discharge status, reasons to stay, and level of care needed), and postdischarge (for example, hospital LOS and destination). Diagnoses on admission and at day 30 were classified according to the ICD-10-CA.²⁹

Analysis

Descriptive statistics including means (standard deviations), medians, proportions, and 95% confidence intervals were used to describe various characteristics. Exploratory univariate analyses, including the chi-squared test or the Fisher exact test compared different factors between the acute and nonacute groups at day 30. The categorization of acute and nonacute status was based on the health care professionals’ evaluations, as specified in the patients’ medical charts. Additional exploratory analyses were conducted to examine factors associated with appropriateness of stay (acute, compared with nonacute) at day 30 of hospitalization. For example, we sought to determine whether living alone or not having family support was associated with inappropriate stay on an acute care unit. All statistical analyses were performed with SAS version 9.3 (SAS Institute, Cary, NC) statistical analysis software.

Results

Pre-Admission

In total, 349 patients were admitted to a psychiatry unit, among which 262 (75%) were identified for the study. More than one-half were male, with a mean age of 45 years (17%; ≥65 years). Twenty-six per cent reported a language other than English or French as their mother tongue, and 7 patients (3%) spoke neither English nor French. Most patients (66.4%) lived at home, 22% lived in a specialized residence or group home, and 11% were homeless. Family was available when needed for 36% of patients, and very few (mainly older patients) were receiving home care from community health services. Fifty-seven per cent were cognitively impaired, of which 81% had a legal representative. Finally, a few patients had special medical needs, such as hearing or visual impairments, nutritional needs, or required a wheelchair (Table 1).

Table 1.

Patient characteristics and situation before the hospital admission

Characteristic	Total n = 262
	n	%
Age, years, mean (SD)	45 (18.0)
Male	149	56.9
Age group, years
18–34	93	35.5
35–49	68	26.0
50–64	58	22.1
≥65	44	16.8
Mother tongue other than English or French	68	26.0
Living arrangements
Home with other adults	93	35.5
Home alone	81	30.9
Residence or group home	57	21.8
Homeless	28	10.7
Jail	3	1.1
Help—family available when needed	94	35.9
Home care	11	4.2
Cognitively impaired	148	56.5
Special medical needs
Hearing impairment	7	2.7
Vision impairment	4	1.5
Nutritional needs (for example, tube feeding)	3	1.1
Wheelchair	2	0.8

On Admission

As shown in Table 2, patients were mainly admitted from the ED. Twenty-seven per cent had only 1 admission diagnosis, 38% had 2 and 35% had 3 or more (maximum of 7). As expected, most patients (99%) had a diagnosis within the category of mental and behavioural disorders (ICD-10-CA; Chapter V).²⁹ In fact, only 4 patients were admitted to a psychiatry unit without an admission diagnosis in the above category. Two patients had a diagnosis in the category factors influencing health status and contact with health services. One patient waiting for a court date was admitted for examination and encounter for administrative purposes, and 1 homeless patient was admitted because of a personal history of risk factors. One patient with a psychotic disorder was admitted for endocrine disorders, and 1 elderly patient with schizophrenia was admitted after a fall. Both were admitted under psychiatry due to a history of mental disorders, and followed by consultants from endocrinology and orthopedics, respectively.

Table 2.

Admission characteristics

Characteristic	Total n = 262
	n	%
Type of admission
From the emergency department	253	96.6
Transfer from another hospital	4	1.5
Direct admission from home	3	1.1
Transfer from jail	2	0.8
Number of admission diagnoses
Only 1 diagnosis	71	27.1
2 diagnoses	100	38.2
≥3 diagnoses (maximum of 7)	91	34.7
Category of admission diagnosisa
Chapter V: Mental and behavioural disorders	258	98.5
F20: Schizophrenia	101	38.5
F29: Unspecified nonorganic psychosis	70	26.7
F25: Schizoaffective disorders	61	23.3
F31: Bipolar affective disorder	55	21.0
F32: Depressive episode	42	16.0
F60: Specific personality disorders	29	11.1
Chapter XXI: Factors influencing health status and contact with health services	91	34.7
Z91: Personal history of risk factors	52	19.8
Z65: Problems related to other psychosocial circumstances	32	12.2
Z86: Personal history of certain other diseases	13	5.0
Chapter VI: Diseases of the nervous system	4	1.5
Chapter IV: Endocrine, nutritional, and metabolic diseases	3	1.1
Chapter XX: External causes of morbidity and mortality	3	1.1
Chapter XIV: Diseases of the genitourinary system	2	0.7
Other (Chapters I, II, III, X, XI, XII, XIII, XIX)	6	2.3

^aNot mutually exclusive

From Admission to Day 30

Seventy-nine consultations (excluding psychiatry) were requested from admission to day 30. In addition to psychiatry consultations, 11% of patients had 1 additional consultation, 4% had 2, 3% had 3, and 1 patient had 4. The 2 most frequent services consulted were internal medicine and endocrinology (Table 3). The most frequent evaluations and treatments were from social services and occupational therapy, respectively. Lastly, 3 patients required infection control isolation and 1 patient spent 3 days in the intensive care unit following a suicide attempt.

Table 3.

Consultations, evaluations, and treatments during the hospital stay

Characteristic	n	%
Types of consultationsa	79
Internal medicine	24	30.4
Endocrinology	16	20.3
Dermatology	7	8.9
Neurology	6	7.6
Gastroenterology	5	6.3
Other medical specialty (cardiology, ENT, geriatrics, intensive care unit, infectious diseases, nephrology, orthopedics, pulmonary, surgery, urology)	21	26.6
Types of evaluationsa	262
Social services	176	67.2
Nutrition	21	8.0
Speech therapy	1	0.4
Thrombosis team	1	0.4
Types of treatmentsa	262
Occupational therapy	137	52.3
Spiritual support group	21	8.0
Psychotherapy	19	7.3
Physiotherapy	17	6.5
Electroconvulsive therapy	17	6.5
Dentistry	7	2.7
Audiology	3	1.1
Radiotherapy	2	0.8
Wound care	1	0.4

^aNot mutually exclusive

ENT = ear, nose, and throat

On Day 30

Based on chart information, 212 patients (81%) still required acute care at day 30 of their hospitalization: 202 patients needed acute psychiatric care, 2 required acute medical care, and 8 patients needed both psychiatric and medical care. Fifty patients (19%; 95% CI 15% to 24%) required nonacute care at day 30: 40 patients were suitable for alternative resources (for example, outpatient follow-up or long-term care) while 10 patients did not require any type of care but remained in hospital for administrative purposes (for example, waiting for a court date or waiting for transfer to jail).

In the group of acute patients at day 30 of their hospitalization (Table 4), 40% were diagnosed with 1 condition requiring acute care, 36% with 2, 21% with 3, and 3% with 4 or more (maximum of 7). Diagnoses were mainly in the category of mental and behavioural disorders. Delusions, hallucinations, inability to take medications independently, and inadequate control of aggression and impulses were the most frequent psychiatric problems present at day 30.

Table 4.

Diagnosis and active psychiatric problems among acute care patients at day 30

Characteristic	Total n = 212
	n	%
Number of diagnoses at day 30
Only 1 diagnosis	85	40.1
2 diagnoses	77	36.3
≥3 diagnoses (maximum of 7)	50	23.6
Category of diagnosis at day 30a
Chapter V: Mental and behavioural disorders	210	99.1
F20: Schizophrenia	67	31.6
F25: Schizoaffective disorders	52	24.5
F29: Unspecified nonorganic psychosis	49	23.1
F31: Bipolar affective disorder	42	19.8
F32: Depressive episode	34	16.0
F60: Specific personality disorders	19	9.0
Chapter XXI: Factors influencing health status and contact with health services	22	10.4
Z91: Personal history of risk factors	8	3.8
Z65: Problems related to other psychosocial circumstances	5	2.4
Chapter XX: External causes of morbidity and mortality	17	8.0
Y84: Other medical procedures as the cause of abnormal reaction of the patient, or later complication, without mention of misadventure at the time of the procedure	17	8.0
Chapter IV: Endocrine, nutritional, and metabolic diseases	12	5.7
Chapter VI: Diseases of the nervous system	6	2.8
Chapter XI: Diseases of the digestive system	6	2.8
Chapter IX: Diseases of the circulatory system	5	2.4
Chapter XIV: Diseases of the genitourinary system	3	1.4
Other (Chapters II, III, X, XII, XIII, XIX)	10	4.7
Active psychiatric problems at day 30a
Delusions and hallucinations significantly interfere with functioning	72	34.0
Inability to take medication independent of hospital or medical staff	50	23.6
Inadequate control over aggression and impulsivity	35	16.5
Inability to carry out basic activities of daily living	10	4.7
Actively suicidal or homicidal	9	4.2

^aNot mutually exclusive

In the group of patients who did not require acute care (Table 5), 38% were diagnosed with 1 condition, 44% with 2, and 18% with 3 or more (maximum of 6). Most patients had a diagnosis in the category of mental and behavioural disorders, followed by factors influencing health status and contact with health services. Among the reasons preventing hospital discharge, difficulty finding or lack of available resources were the most frequent (for example, waiting for ambulatory care, rehabilitation, or long-term care placement). Other reasons included administrative and (or) social issues (for example, waiting for a court date or waiting transfer to jail) and ongoing assessment to determine the appropriate resources.

Table 5.

Diagnosis and reasons to stay in hospital among nonacute care patients at day 30

Characteristic	Total n = 50
	n	%
Number of diagnoses at day 30
Only 1 diagnosis	19	38.0
2 diagnoses	22	44.0
≥3 diagnoses (maximum of 6)	9	18.0
Category of diagnosis at day 30a
Chapter V: Mental and behavioural disorders	46	92.0
F20: Schizophrenia	21	42.0
F25: Schizoaffective disorders	12	24.0
F31: Bipolar affective disorder	8	16.0
F29: Unspecified nonorganic psychosis	6	12.0
F32: Depressive episode	4	8.0
Chapter XXI: Factors influencing health status and contact with health services	25	50.0
Z75: Problems related to medical facilities and other health care	12	24.0
Z02: Examination and encounter for administrative purposes	8	16.0
Other (Chapters VI, XIII, XIV, XX)	5	10.0
Reasons for no discharge at day 30
Appropriate resources lacking or difficult to find	33	66.0
Administrative or social issues	15	30.0
No longer acute but ongoing assessment to determine appropriate resources	14	28.0
Ongoing liaison process with community care staff	6	12.0
Ongoing family discussion	4	8.0
Waiting for a specific treatment	3	6.0

^aNot mutually exclusive

Postdischarge

At the end of our study (June 2011), hospital LOS and discharge destination information was obtained for all patients. Almost all patients living at home (173/175; median LOS = 45 days/maximum of 298 days) or in a group home or specialized residence (55/56; median LOS = 54 days/maximum of 318 days) returned to their original residences. One patient living at home was transferred to another acute care hospital after 129 days, while another patient previously living in a group home was relocated to a long-term care institution for nonautonomous patients after 92 days. As for homeless patients, they were all relocated (27/28; median LOS = 53 days/maximum of 167 days) except for 1 patient who left the hospital against medical advice after 108 days. For patients incarcerated at the time of admission, 1 was transferred back to jail after 161 days and 2 patients were released by the court and returned home: 1 after 84 days, the other after 507 days. Finally, 1 patient living at home who was admitted to the hospital on June 1, 2009 was still hospitalized, 2 years later.

Exploratory Analyses

Univariate analyses did not reveal significant factors associated with nonacute status (most likely due to the small sample of nonacute patients [n = 50]). However, although not statistically significant, the proportion of nonacute patients was higher for those with fewer admitting diagnoses (≤2 = 23%, compared with >3 = 13%) and lack of family support (no support = 23%, compared with support = 13%).

Discussion

Our study describes psychiatric patients occupying acute care beds in psychiatric units and based on their needs, determines the level of care required (acute, compared with nonacute) at day 30 of their hospitalization.

At day 30 of hospitalization, 19% of patients occupied an acute care bed, despite being nonacute. This is consistent with other studies confirming the use of acute care beds for nonacute needs.^19,20 Various reasons preventing these patients from leaving the hospital were identified. The most frequent was the lack of, or difficulties in finding, appropriate resources that best met the patients’ needs. In fact, previous work has shown that the demand for alternative services (for example, long-term care, community support, or rehabilitation) often exceeds the available resources.^1,17,30,31 The complexity of patients’ medical needs increases the difficulty in finding appropriate resources in a timely manner.

As a result, rehousing or permanent placement of patients with multiple comorbidities (for example, mental retardation, schizophrenic patients with violent behaviour, or patients with physical limitations or handicaps suffering from psychiatric disease) is associated with long discharge delays.^32,33 In such cases, collaboration between all relevant in- and outpatient services is crucial in facilitating the discharge process.³⁴

As well, access to subacute facilities that provide a level of care appropriate for specific clienteles (for example, elderly patients with psychiatric illnesses) could be difficult.³⁵ Studies have shown that subacute facilities, such as nursing homes, are rarely equipped or staffed to provide appropriate services to patients with psychiatric conditions.^36,37 Further, the number of other placement options capable of providing optimal care (that is, rehabilitation, home care, and [or] community services) is insufficient to adequately respond to the demand.^1,17,30,31 This, in part, explains why psychiatric patients remain in hospital longer, compared with nonpsychiatric patients.³⁰ Moreover, discharge options are sometimes limited for patients with multiple medical and psychiatric comorbidities, who have complex needs. As a result, this patient population is at greater risk for occupying an acute care bed, even after the acute episode has been managed.^19,30 Paton et al¹⁹ found that an acute psychiatric care ward was not an appropriate setting for up to 40% of elderly patients hospitalized on that unit. For optimal resource use, as well as to address patients’ needs in the best possible manner, all alternatives to acute psychiatric care units, including community support and (or) residential care, should be considered.³⁸

Our study also demonstrated that over 80% of patients continued to require acute psychiatric care at day 30 of their hospitalization, while only 5% required both psychiatric and medical acute care. Availability and awareness of alternative resources for patients with acute psychiatric needs would also allow for the identification of patients potentially suitable for these resources. Previous research has shown that alternative resources could be beneficial for patients who require hospital level acute psychiatric care,^39–41 without medical care.⁴² For example, in our study, 24% of patients were unable to take their medication independently and 5% were unable to carry out basic activities of daily living. As these types of care could be provided in subacute settings, such as residences or nursing homes, access to such resources would potentially facilitate discharging these patients. Lastly, although some alternative resources can be provided to patients with acute psychiatric care needs, those with serious problems, such as active suicidal or homicidal behaviour would be more appropriate for hospital-based programs.^43,44

Improved access and use of alternatives to acute psychiatric beds for patients suitable for subacute facilities would also benefit EDs. Psychiatric patients represent a major challenge for EDs,¹ mainly because of their increased volume,^2,8,45,46 long ED LOS,^2,11,47,48 and high admission rates.^8,10 In our study, 97% of our patient population was admitted from the ED, with an average LOS of 37 and 76 hours for the JGH and NDH, respectively. Increasing access to inpatient beds would undoubtedly reduce the burden on EDs¹ by reducing the boarding times of admitted patients. Similarly, quality of care would be improved as patients would be managed by specialized psychiatric personnel instead of ED staff, who are not adequately trained to deal effectively with the specific needs of this clientele.^1,3,49

More than one-half of our patients were cognitively impaired; a frequent occurrence among patients with mental disorders.⁵⁰ This characteristic has been found to be associated with longer hospitalization.^51,52 Cognitively impaired patients remaining on an acute psychiatric care unit present a major concern, mainly because these settings are often unable to adequately meet these patients’ needs.^53,54

Finally, 8 patients in the nonacute group had a total hospital LOS of greater than 100 days, with 1 patient staying as long as 153 days. For the longest hospitalizations, a thorough chart review revealed 2 main reasons explaining the prolonged stays. First, the appropriate resources were very difficult to find for patients with multiple medical and psychiatric needs (for example, elderly handicapped patients suffering from schizophrenia, suicidal patients with drug addiction, or homeless patients with psychosis and with criminal records). This result corresponds with other studies^18–20,35 that have reported limited access to alternative resources for patients with multiple problems. Second, in the case of patients admitted under a court decision (for example, examination and encounter for administrative purposes or direct transfer from jail), delays were incurred due to provincial judicial administrative backlogs^55,56 out of the control of the hospitals.

Our study has some limitations. First, the designation of acute, compared with nonacute, status was subjectively determined based on the medical team’s opinions. Second, our study’s time frame did not allow for the identification of factors related to inappropriate use of acute care beds (mainly due to the small sample; n = 50 nonacute patients). Further, given that day 30 was the point of interest selected in our study, there is the possibility that nonacute patients discharged prior could have exhibited unique characteristics that were not captured in our study. Lastly, applicability of our findings to other health care environments may be limited because of differences in how health care systems are organized and funded outside Canada.

Nevertheless, our study provides relevant data that will be useful in designing future research. For example, for studies designed to identify risk factors for nonacute prolonged hospital stays or those related to examining the financial impact of nonacute hospital stays, compared with other alternative settings.

Conclusion

Our study demonstrates that over 80% of patients continued to require acute psychiatric care at day 30 of their hospitalization, while 19% occupied an acute care bed despite their acute episode having been resolved. Delusions, hallucinations, inability to take medications independently, and inadequate control over aggression and impulses were the most common reasons requiring acute psychiatric care at day 30. A certain number of patients currently designated acute care may benefit from alternative resources to being hospitalized on an acute care psychiatric unit.

The results of our study highlight the challenges and the unrealistic expectations faced by psychiatric acute care units. For example, resources destined for the aging population are often inadequate for psychiatric patients and it is not unusual that long-term care facilities refuse to accept patients with psychiatric disorders (for example, elderly patients suffering from schizophrenia are very difficult to place when their condition prevents them from being discharged home).

When these patients cannot be discharged to alternative settings for whatever reason (resource unavailable, administrative, or social issues), acute care units become the holding area. These challenges need to be proactively addressed, potentially through the development of policies, interventions, and research focused on coordination of specialized services with primary care and improved intersectorial collaborations in the organization of mental health service.

Abbreviations

ED

emergency department

ICD

International Classification of Diseases

JGH

Jewish General Hospital

LOS

length of stay

NDH

Notre-Dame Hospital